Web Service Selection Using Modified Artificial Bee Colony Algorithm

Web services are a type of application software, which can be remotely accessed through the Internet. Due to the proliferating growth of web services of the same functionality, the user goes into a dilemma to select suitable service for him. In this paper, we study the web service selection (WSS) problem in a sequential composition model. We formulated the WSS as a constrained optimization problem. To solve the problem, we suggest a modified artificial bee colony (mABC) algorithm, which uses a chaotic-based opposition learning method to generate a better initial population. To improve the exploration capability of the mABC, a new search equation for employed bee phase is suggested. On the other hand, to improve the exploitation capacity of the mABC, a new search strategy, inspired by differential evolution (DE), is adopted in the onlooker bee phase. We test the mABC on synthetic web service selection problem taken from QWS dataset. To assess the relative performance of the mABC, we compare it against five other state-of-the-art algorithms. The experimental results show that the mABC is better than other existing approaches in terms of response time, latency, availability, and reliability.